Four assumptions of multiple regression that researchers should always test

Osbourne, Jason W.; Waters, Elaine

doi:10.7275/r222-hv23

Cited by 158 publications

(66 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We calculated standardized z-scores to identify outlying data, using the cut-off of values of |z|> 3. Traditionally, for nonnormal data, transformations have been recommended to approximate a normal distribution in prior research [ 108 – 110 ], since transformations can bring about normality in the data, minimize the influence of extreme values on the dataset, and potentially improve the power of parametric analyses [ 111 , 112 ]. However, a drawback of transformations is the increased difficulty of interpretation of the relationships between the independent and dependent variables [ 113 ].…”

Section: Methodsmentioning

confidence: 99%

The Acceptance of Telepresence Robots in Higher Education

Lei

Clemente

Liu

et al. 2022

Int J of Soc Robotics

View full text Add to dashboard Cite

While telepresence robots have increasingly become accepted in diverse settings, the research on their acceptance in educational contexts has been underdeveloped. This study analyzed how the use intention of telepresence robots can be influenced by perceived usefulness, perceived ease of use, subjective norm, and perceived risk for students, faculty, and staff in higher education. Survey data were collected from 60 participants with direct operator experience with a variety of telepresence robots deployed in a large research university in the Midwest region of the United States. Path analysis results indicated that perceived usefulness was the only significant direct predictor of use intention of telepresence robots. Both perceived ease of use and subjective norm had a significant positive effect on perceived usefulness. Subjective norm also had a significant positive indirect effect on use intention, mediated by perceived usefulness. Perceived risk had a negative effect on perceived ease of use. These findings indicated that the usefulness of robots was central to operators’ decisions to use telepresence robots. Therefore, design choice for telepresence robots should prioritize usefulness. Secondly, the design of telepresence robots should minimize complexity for the end user and minimize cognitive demand. Having nominal difficulty of use would also facilitate multiple embodiments by combining telepresence robots with other technologies to support more rich social interactions.

show abstract

Section: Methodsmentioning

confidence: 99%

The Acceptance of Telepresence Robots in Higher Education

Lei

Clemente

Liu

et al. 2022

Int J of Soc Robotics

View full text Add to dashboard Cite

show abstract

“…In this study we collected, variables based on discussions in the research literature, and performed analyses to see whether parking behaviour is significantly affected by these variables. In order to obtain reliable results from the MLR, a number of tests and adjustments are required, including linearization of dependent and independent variables, and testing for multicollinearity and for homoscedasticity of error variances (Osborne and Waters, 2002;Uyanık and Güler, 2013) using the Breusch-Pagan test. For estimation of variables, the variables need to be tested for their relevance, using R-square values for comparing predicted and actual values.…”

Section: Methods and Datamentioning

confidence: 99%

Towards Sustainable Urban Car–Parking Solutions: Exploring Effects of Parking Policies Using Spatial Regression Analysis

Tsai¹,

Gasselt²,

Lin³

2021

giforum

View full text Add to dashboard Cite

In urban policy development and transport planning, car parking has become a crucial topic, as an ill-designed system could lead to traffic congestion, safety problems, increased air pollution and other challenges, particularly in densely populated urban areas. Adjusting parking fees has been deemed the most effective tool to control parking behaviour for selected locations, in order to attract or repel drivers. However, in order to select target areas efficiently, and before any recommendation for fee adjustment is made, an analysis of the current parking situation and the identification of variables influencing parking behaviour are needed.We present a spatial analysis targeted at parking demand and supply data gathered via road surveys by local authorities in Taipei City. The spatial analysis is complemented by a regression analysis to identify variables that influence parking behaviour.Our approach examines the relationship between potential controlling and contributing factors, and we show the influence of variables in a specific example. Our aim is to provide a starting point for future policy development and pricing adjustment at a local level. This initial framework might provide a conceptual core for wider discussion and a tool for integrating other different scenarios in the future.

show abstract

“…Visual inspection of plots with such deviations is relevant when checking for the assumption of homoscedasticity in regressions, which means that the variance of errors is the same across all levels of the independent variable (Osborne & Waters, 2002).…”

Section: Methodsmentioning

confidence: 99%

Visual Model Fit Estimation in Scatterplots and Distribution of Attention

Reimann

Blech

Gaschler

2020

Experimental Psychology

View full text Add to dashboard Cite

Abstract. Scatterplots are ubiquitous data graphs and can be used to depict how well data fit to a quantitative theory. We investigated which information is used for such estimates. In Experiment 1 ( N = 25), we tested the influence of slope and noise on perceived fit between a linear model and data points. Additionally, eye tracking was used to analyze the deployment of attention. Visual fit estimation might mimic one or the other statistical estimate: If participants were influenced by noise only, this would suggest that their subjective judgment was similar to root mean square error. If slope was relevant, subjective estimation would mimic variance explained. While the influence of noise on estimated fit was stronger, we also found an influence of slope. As most of the fixations fell into the center of the scatterplot, in Experiment 2 ( N = 51), we tested whether location of noise affects judgment. Indeed, high noise influenced the judgment of fit more strongly if it was located in the middle of the scatterplot. Visual fit estimates seem to be driven by the center of the scatterplot and to mimic variance explained.

show abstract

Four assumptions of multiple regression that researchers should always test

Cited by 158 publications

References 4 publications

The Acceptance of Telepresence Robots in Higher Education

The Acceptance of Telepresence Robots in Higher Education

Towards Sustainable Urban Car–Parking Solutions: Exploring Effects of Parking Policies Using Spatial Regression Analysis

Visual Model Fit Estimation in Scatterplots and Distribution of Attention

Contact Info

Product

Resources

About